A New Metric for the Assessment of Driver Crash Risks

Road accidents account for thousands of deaths and injuries each year in Great Britain. Central to concerns for road safety are younger and older drivers who are reported to have crash rates per mile driven or per trip made that are much greater than drivers of other ages. Policymakers in Great Britain and around the world have responded to the high crash rates of these two groups of drivers by proposing initiatives to reduce their exposure to risk. Graduated licensing systems in the United States and other countries impose restrictions on young drivers (e.g., carrying passengers, curfews). The Department for Transport has recommended similar graduated licensing in Great Britain. At present, license renewal regulations are imposed on older drivers to identify visual, cognitive, and physical impairment among older adults that might put them at increased risk as drivers.

This research project will investigate a bias in driver crash risks that potentially exaggerates the risks of young and elderly drivers, leading to inaccurate reports of driver crash risks as well as misinforming road safety policies designed to safeguard road users. We will develop a new metric for the assessment of driver crash risk that overcomes previous bias with a view to better informing road safety policies and the general public. Public misconceptions about driver crash risks impact on people's lives such as by discouraging older drivers from renewing their driver license, which directly affects their mobility, health, and well-being.

Shared beliefs and misconceptions about the risks of young and elderly drivers have the potential to influence eye-witness accounts and police officers' reporting of road accidents, especially with regard to the allocation of driver blame. Age bias in police reporting has direct implications for the reliability of police reported accident data on which estimates of crash risk are based. This project investigates the potential for police reporting bias with a series of psychological studies that identify situations where reporting bias is most likely to occur, enabling recommendations to be made for how such bias can be reduced in road accident assessment.

Policymakers are charged with the difficult role of designing policies to safeguard the public. With regards to road safety, these policies must balance the benefits of driving with a need for public safety. For example, license renewal regulation is effective for older drivers who are impaired and may safeguard them from potential injury. However, an unintended outcome of license renewal regulation has been to discourage older drivers from renewing their driving license, which compromises their mobility with negative consequences for their health and well-being. It is extremely important then that such policies are informed by accurate assessments of driver risk. The outputs of this research will impact on policymakers by providing reliable and accurate assessments of driver risk, which will enable them to allocate resources appropriately to safeguard road users.

For all police reported crashes in Great Britain the investigating officer records factors that they believe may have contributed. These factors include driver error (e.g., failure to look properly) and impairment (e.g., alcohol consumption) among others. The reporting officer may also draw on reports of witnesses in the general public. Accident reports are subjective judgments and are susceptible to bias. By identifying situations in which bias is most likely to occur, this project will inform police officers about how they can reduce the risk of bias in their reporting of road accidents. In turn, this will improve the accuracy and reliability of driver risk estimates that are based on police reported data.

Vehicle insurers rely on estimates of potential loss when identifying the appropriate insurance cost for a driver. Driver age is central to these assessments and young drivers in particular are charged a high premium for vehicle insurance. Insurers often rely on their own risk assessments. This project will identify bias in previous estimates of driver crash risk that will have influenced insurers own risk assessments. The outputs of this research will impact on vehicle insurers by providing a more accurate tool for assessing driver crash risk. The project also investigates potential bias in police and witness reports of road accidents. Vehicle insurers rely on the accuracy of these reports in their evaluation of insurance claims. This project will inform insurers about the situations in which bias is most likely to occur.

Investigating bias in crash rate estimation and police practice in reporting of road accidents, this project will impact on road safety policymaking, vehicle insurer practice, and the society in general through improvements in public health and well-being.